We describe the construction of a paper-and-pencil Test of Basic Scientific Literacy (TBSL) specifically designed for high-school leavers entering technikon and university in South Africa. The `true-false-don't know' scientific literacy test-items are based on a pool of 472 items developed previously from selected literacy goals recommended by the AAAS in Science for all Americans. Test-items were pilot-tested on 625 technikon and university students and were included in the 110-item TBSL on the basis of item discrimination, item difficulty and student feedback. The TBSL consists of three subtests based on Jon Miller's three constitutive dimensions of scientific literacy: the nature of science (22 items); science content knowledge (72 items); and the impact of science and technology on society (16 items). About 260 South Africa-based members of various South African professional science and engineering associations participated in setting a performance standard for each of the three dimensions of scientific literacy. These standards were validated using a `contrasting groups' approach. The internal consistency of the individual TBSL subtests, and the reliability of mastery-nonmastery classification decisions based on the performance standard, was found to be about 0.80. The reliability of the overall 110-item TBSL was 0.95.

We describe the development of a pool of paper‐and‐pencil test‐items of scientific literacy that are based on selected chapters of the 1989 American Association for the Advancement of Science overview report on literacy goals in science, mathematics, and technology, entitled Science for All Americans. Separate pools of test‐items were developed for each of Miller's “three constitutive dimensions” of scientific literacy. Through carefully structured panel reviews using Fellows of the Royal Society of South Africa and university lecturers, we have developed 472 true—false test‐items with high content and item validity, as well as wide applicability and use, that test 240 key ideas in, and attitudes toward, science. These items cover the earth, physical/chemical, life and health sciences, as well as the nature of science and the impact of science and technology on society. We have taken an essentially complementary approach to Miller's research methodology, and believe that these test‐items represent a refinement of previous attempts to measure the three dimensions of scientific literacy. © 1996 John Wiley & Sons, Inc.

No abstract

The intuitive beliefs about a number of well-known physical events were investigated in a longitudinal and cross-cultural study involving 2326 pupils in schools in the Cape Province of the Republic of South Africa and the then Republic of Transkei using a multiple-choice questionnaire designed for the purpose. One of the findings of the study is that the context of the question has a clear effect on the frequency with which different options are selected in physically similar situations. Furthermore, differences between the beliefs of pupils in schools in the Cape and in Transkei emerge, as well as an increasing trend towards the selection of the correct response in some situations across the standards in Cape schools which appear to indicate the effect of learning in science.

No abstract